J. clin. Path., 1975, 28, 775-778
A comparison of the in vitro activity of metronidazole, tinidazole, and nimorazole against Gram-negative anaerobic bacilli A. V. REYNOLDS, J. M. T. HAMILTON-MILLER, AND W. BRUMFITT From the Royal Free Hospital, Department of Medical Microbiology, Pond Street, Hampstead, London NW3 2QG
The in vitro activities of metronidazole, nimorazole, and tinidazole were compared against 69 strains of obligately anaerobic Gram-negative bacilli. Geometric mean MICs were 0 34, 1 05, and 0-28 Hg/ml respectively. Thirty-six strains were also tested by the disk method. Correlation between MIC and diameter of the zones of inhibition was poor.
SYNOPSIS
Metronidazole (1-,B-hydroxymethyl-2-methyl-5nitroimidazole, Flagyl, May & Baker) is at present the drug used most widely for the treatment of Trichomonas vaginalis infection. It is of interest that it has also been effective in the treatment of acute ulcerative gingivitis (Shinn, 1962; Davies et al, 1964; Shinn et al, 1965), an infection in which anaerobic bacteria (Treponema vincentii and Fusobacterium spp.) are believed to have an aetiological r6le. Recent studies have shown that all the clinically important species of anaerobic bacteria are sensitive to therapeutically attainable concentrations of metronidazole (Fuzi and Csukas, 1969a, b, c, 1970; Prince et al, 1969; Ueno et al, 1971; Videau, 1971; Nastro and Finegold, 1972; Tally et al, 1972, 1974; Finegold and Sutter, 1972; Whelan and Hale, 1973; Mitchell, 1973; Washington et al, 1974; Dornbusch and Nord, 1974; Phillips, 1974; Staneck and Washington, 1974). In addition, metronidazole is consistently bactericidal in action, which is a property not shown by clindamycin, rifampicin or chloramphenicol (Nastro and Finegold, 1972; Finegold and Sutter, 1972; Whelan and Hale, 1973). Two reports have appeared on the clinical use of metronidazole: Tally et al (1972) successfully treated three patients with infections caused by Gramnegative anaerobic bacteria, and the Study Group (1974) showed that metronidazole was effective both therapeutically and prophylactically in gynaecological patients with pelvic sepsis. Nimorazole1 (1-(N-P-ethylmorpholino)-5-nitro"Previously known as nitrinidazine. Received for publication 14 April 1975.
imidazole, Nagoxin, Carlo Erba) and tinidazole (1
(2-ethylsulphonylethyl)-2-methyl-5-nitroimidazole Fasigyn, Pfizer) are also 5-nitroimidazoles whose antitrichomonal activity in vitro is similar to that of metronidazole (de Carneri et al, 1969; Howes et al, 1970). Both have been used successfully to treat trichomoniasis, with reduction in both daily doses and treatment period compared with metronidazole (Rognoni and Sideri, 1969; Gyorik and Wenner, 1971). Nimorazole was also found to be as effective as metronidazole in the treatment of acute ulcerative gingivitis (Lozdan et al, 1971). Considering the superior pharmacokinetic properties of nimorazole and tinidazole, we thought it worthwhile to carry out a comparison of the in vitro activity of these three 5-nitroimidazoles against Gram-negative anaerobic bacteria. Methods and Materials
5-NITROIMIDAZOLES Pure samples of metronidazole, nimorazole, and tinidazole were supplied by May & Baker, Carlo Erba, and Pfizer, respectively. BACTERIAL STRAINS
Sixty-nine strains were studied: all were obligately anaerobic Gram-negative rods. Twenty-four strains (8 Fusobacterium spp. and 16 Bacteroides fragilis) were either reference strains (from the National Collection of Type Cultures, London or Center for Disease Control, Ga, USA) or identified clinical isolates from other hospitals. The majority of the remaining 45 strains had been isolated from clinical 775
A. V. Reynolds, J. M. T. Hamilton-Miller, and W. Brumfitt
776
material from the Royal Free Hospital but some were kindly donated by other hospitals. We provisionally identified 43 of these as B. fragilis, since they were resistant to polymyxin B (10 ,ug/ml), kanamycin (2 mg/ml), and benzylpenicillin (2 u/ml) (Finegold et al, 1967). The other two strains were sensitive to benzylpenicillin and thus were classified as Bacteroides spp. MIC DETERMINATION
Throughout these experiments, media were supplemented with yeast extract (1 %), haemin (5 ,ug/ml), and menadione (0 5 ,ug/ml). Serial doubling dilutions of the drugs were made in brain heart infusion agar (Oxoid, CM 375) containing 5 % lysed, defibrinated horse blood. Bacteria were grown in 10 ml amounts of thioglycollate broth (Baltimore Biological Laboratories, no. 135C) incubated anaerobically at 37°C for 22 hr, using the GasPak system. Such cultures were diluted 1:500 in sterile 0-06 M sodium phosphate buffer containing 0 3 g/l cysteine hydrochloride, and inoculated on to the agar with a multiple inoculating device (see Reynolds et al, 1974). Plates were incubated at 37°C for 42 hr using the GasPak system. Control plates containing no antibiotic were inoculated and incubated both aerobically and anaerobically. MIC was taken as the lowest drug concentration that completely inhibited growth or MIC
allowed the development of no more than two discrete colonies; such a finding is evidence of at least 99 95 % inhibition. Disk Diffusion Tests
Broth cultures were grown as above and diluted 1: 50 in buffer. Fifteen millilitre amounts of supplemented (see above) brain heart infusion agar + 5 % whole horse blood in 9 cm plates were flooded with the diluted cultures and excess fluid was removed. Disks containing 2f5 ,ug of drug were prepared by dipping 6 mm AA disks (Whatman) into solutions of each individual drug which contained 125 ,g/ml. These were then placed onto the surface of the agar. Only one disk was put on to each plate. Plates were incubated anaerobically using the GasPak system for 18 hr at 37°C, and zones of inhibition were measured with calipers. Analysis of Data Graphs were constructed to compare zone size with MIC for each compound; in addition, results for all three compounds were combined onto a single plot. Analysis was as described by Ericsson and Sherris (1971). Results
As
can
be
seen
from table I, there
was no
(Mg/ml) 2-0
0 062
0 125
0-25
05
1
2 3
4 27 1
24 1
3
7
2 41 2
5 6
1 15 1
3
1
1.0
40
8-0
Metronidazole
Fusobacterium sp B. fragilis Bacteroides sp Nimorazole Fusobacterium sp B. fragilis Bacteroides sp Tinidazole Fusobacterium sp B. fragilis Bacteroides sp
5 31 1 Metronidazole 0 34 ug/ml Nimorazole 1-05 Mg/ml 0-28 pg/ml Tinidazole 2 7
2
Geometric mean MIC
1 4
1 1 1
Table I Activity of metronidazole, nimorazole, and tinidazole against Gram-negative anaerobic bacteria Nitroimidazole
Fitted Line y a
Metromidazole
-2-23
Nimorazole Tinidazole
- 2-84 - 162 - 2-90
Composite
=
Correlation Coefficient
SE of Estimate
0 59 0-63 0 53 0 75
3-33 3-62 2 71 3 49
ax + b
b 41-45 43-67 36-46 45-41
Table II Correlation of MIC to zone diameter
1
clear
Nitroimidazoles against Gram-negative anaerobic bacilli
777
difference between the sensitivity patterns of Fusobacterium spp and B. fragilis; therefore it was considered to be justified to combine data for all strains. Tinidazole was slightly more active than was metronidazole, the two compounds having geometric mean MICs of 0-28 and 0 34 ,ug/ml respectively. Nimorazole was about threefold less active, having a geometric mean MIC of 1 05 .tg/ml. Thirty-six strains were also tested by the disk method, and the results are summarized in table IL. To increase the spread of values and thereby make the interpretation of results easier a composite scatter diagram was constructed and a statistical analysis of this is also shown in table II.
nitrofurans. The possession of such a group may be the determining factor in such activity, and it will be of interest to test this hypothesis with further compounds. The nitroimidazoles are at present available only as oral preparations, which may be a disadvantage if the treatment of critically ill patients is contemplated. In view of the fact that metronidazole has been used successfully on a small scale for the treatment of systemic bacterial infections, the superior pharmacokinetics (Welling and Monro, 1972; de Carneri, 1969) of the two other related compounds allied to their equivalent antimicrobial activity make clinical studies of tinidazole and nimorazole mandatory.
Discussion
We should like to thank Dr McFadzean, Dr Rooney, and Dr O'Neill East for the gifts of metronidazole, nimorazole, and tinidazole respectively. Doctors S. M. Finegold and V. L. Sutter, Wadsworth VA Hospital, Los Angeles, California, H. Beerens, University of Lille, France, M. Sebald, Pasteur Institute, Paris, France, G. Lombard, Center for Disease Control, Atlanta, Georgia, R. B. Sykes, Glaxo Research, Greenford, Middlesex, B. S. Drasar, St Mary's Hospital, London, and A. Percival, University of Liverpool, kindly provided strains of Gram-negative anaerobes.
A large number of reports over the past few years have shown metronidazole to be highly active against both Gram-positive and Gram-negative anaerobic bacteria (for review see Hamilton-Miller, 1975). Two other nitroimidazoles, nimorazole and tinidazole, are currently available for the treatment of trichomoniasis (although tinidazole is not at present available on prescription in the UK.) We have not seen any reports on the antibacterial activity of nimorazole, and only two concerning tinidazole. Edwards et al (1973) reported that tinidazole was slightly less active than metronidazole against eight strains of Clostridium spp, a conclusion confirmed by Dornbusch and Nord (1974), who also found tinidazole to be more active than metronidazole against Gram-negative organisms. Our results have been extended to include nimorazole but in terms of metronidazole and tinidazole they agree very closely with those of Dornbusch and Nord (1974). Summation of published data available on the activity of metronidazole against anaerobic bacteria shows that 1052 out of 1071 strains were sensitive (taking the break-point as being either 12-5 or 16 ,ug/ml). Poor correlation was found in the disk test between zone size and sensitivity. However, since all the strains tested here may be regarded as sensitive, criteria allowing precise interpretation cannot be stated. We suggest an MIC of 16 ,ug/ml as the breakpoint, since this value should be obtainable systemically after a 1 g dose of any of these three compounds. If this value is accepted, strains giving a zone size greater than 13 mm for metronidazole, 7 mm for nimorazole, and 15 mm for tinidazole can be considered as sensitive. Three groups of compounds containing a primary nitro group among other features have now been shown to possess good activity against anaerobic bacteria, chloramphenicol, nitroimidazoles, and
References
Davies, A. H., McFadzean, J. A., and Squires, S. (1964). Treatment of Vincent's stomatitis with metronidazole. Brit. med. J., 1, 1149-1150. de Carneri, 1. (1969). Antiprotozoan activity of nitroimidazoles. Arzneimittel Forsch., 19, 382-386. de Carneri, I., Cantone, A., Emaneuli, A., Giraldi, P. N., Logemann, W., Meinardi, G., Monti, G., Nannini, G., Tosolini, G., and Vita, G. (1969). Nitrimidazine: a new systemic trichomonacide. 6th International Congress of Chemotherapy, Tokyo (Kyoto). Dornbusch, K. and Nord, C. E. (1974). In vitro effect of metronidazole and tinidazole on anaerobic bacteria. Med. microbiol. Immunol., 160, 265-267. Edwards, D. I., Dye, M., and Came, H. (1973). The selective toxicity of antimicrobial nitroheterocyclic drugs. J. gen. Microbiol., 76, 135-145. Ericsson, H. M. and Sherris, J. C. (1971). Antibiotic sensitivity testing. Report of an international collaborative study. Acta. path. microbiol. scand. Section B., Suppl. no. 217.
Finegold, S. M., Harada, N. E., and Miller, L. G. (1967). Antibiotic susceptibility patterns as aids in classification and characterization of gram-negative anaerobic bacilli. J. Bact., 94, 1443-1450. Finegold, S. M. and Sutter, V. L. (1972). Antimicrobial susceptibility of anaerobic Gram-negative bacilli. In Host Resistance to Commensal Bacteria, edited by T. MacPhee. Churchill Livingstone, Edinburgh. Fuzi, M. and Csukas, Z. (1969a). A metronidazol egy eddig ismeretlen antibacterialis hatasa. Orv. Hetil., 110, 16051606.
Fuzi, M. and Csukas, Z. (1969b). A szAjfl6ra mikroorganiz-
778
A. V. Reynolds, J. M. T. Hamilton-Miller, and W. Brumfitt
musainak metronidazol erzekenysege. Orv. Hetil., 110, 2154:2155. Fuzi, M. and Csukas, Z. (1969c). Veillonella-tdrzsek metronidazol-erzekenysege. Fogorv. Szle., 62, 324-326. Fuzi, M. and Csukas, Z. (1970). Das antibacterielle Wirkungs-spektrum des Metronidazols. Zbl. Bakt. I. Abt. Orig., 213, 258-262. Gyorik, W. and Wenner, R. (1971). Therapie der Trichomonaden-infektion mit Tinidazol im Vergleich zu Metronidazol. Schweiz. Rundschau. Med. (PRAXIS), 60, 1612-1614. Hamilton-Miller, J. M. T. (1975). Antimicrobial agents active against anaerobes. J. antimicrob. Chemother., 1 (in press). Howes, H. L. Jr., Lynch, J. E., and Kivlin, J. L. (1970). Tinidazole, a new antiprotozoal agent: effect on Trichomonas and other protozoa. Antimicrobial Agents and Chemotherapy-1969, 8, 261-266. Lozdan, J., Sheiham, A., Pearlman, B. A., Keiser, B., Rachanis, C. C., and Meyer, R. (1971). The use of nitrimidazine in the treatment of acute ulcerative gingivitis. A double-blind controlled trial. Brit. dent. J., 130, 294-296. Mitchell, A. A. B. (1973). Incidence and isolation of Bacteroides species from clinical material and their sensitivity to antibiotics. J. clin. Path., 26, 738-741. Nastro, L. J. and Finegold, S. M. (1972). Bactericidal activity of five antimicrobial agents against Bacteroides fragilis. J. infect. Dis., 126, 104-107. Phillips, 1. (1974). Antibiotic sensitivity of non-sporing anaerobes. In Infections with Non-Sporing Anaerobic Bacteria, edited by I. Phillips, and M. Sussman. Churchill, Livingstone, Edinburgh. Prince, H. M., Grunberg, E., Titsworth, E., and deLorenzo, W. F. (1969). Effects of 1-(2-nitro-1-imidazolyl)-3-methoxy -2-propanol and 2-methyl-5-nitroimidazole-1-ethanol against anaerobic and aerobic bacteria and protozoa. Appl. Microbiol., 18, 728-730. Reynolds, A. V., Hamilton-Miller, J. M. T., and Brumfitt, W. (1974). Newer aminoglycosides-amikacin and tobramycin: an in-vitro comparison with kanamycin and gentamicin. Brit. med. J., 3, 778-780.
Rognoni, V. and Sideri, L. (1969). Treatment of vaginitis with particular reference to those forms due to a protozoan causative factor. Clinical results with a new nitroimidazole preparation. Riv. Ostet. Ginec., 51, 237-246. Shinn, D. L. S. (1962). Metronidazole in acute ulcerative gingivitis. Lancet, 1, 1191. Shinn, D. L. S., Squires, S., and McFadzean, J. A. (1965). The treatment of Vincent's disease with metronidazole. Dent. Practit. dent. Rec., 15, 275-280. Staneck, J. L. and Washington, J. A. 1I. (1974). Antimicrobial susceptibilities of anaerobic bacteria: recent clinical isolates. Antimicrob. Agents Chemother., 6, 311315. Study Group (1974). Metronidazole in the prevention and treatment of bacteroides infections in gynaecological patients. Lancet, 2, 1540-1543. Tally, F. P., Armfield, A. Y., Dowell, V. R., Jr., Kwok, Y-Y., Sutter, V. L., and Finegold, S. M. (1974). Susceptibility of Clostridiumramosum to antimicrobial agents. Antimicrob. Agents Chemother., 5, 589-593. Tally, F. P., Sutter, V. L., and Finegold, S. M. (1972). Metronidazole versus anaerobes. In vitro data and initial clinical observations. Calif. Med., 117, no. 6, 22-26. Ueno, K., Ninomiya, K., and Suzuki, S. (1971). Antibacterial activity of metronidazole against anaerobic bacteria. Chemotherapy (Japan), 19, 111-114. Videau, D. (1971). Association metronidazole-spiramycin sur les germes ana6robes. Path. Biol., 19, 661-666. Washington, J. A., II, Martin, W. J., and Hermans, P. E. (1974). In vitro susceptibility of anaerobic bacteria isolated from blood cultures. In Anaerobic Bacteria: Role in Disease, p. 427, edited by A. Balows, R. M. Dehaan, V. R. Dowell, Jr., and L. B. Guze. Thomas, Springfield, Illinois. Welling P. G. and Monro, A. M. (1972). The pharmacokinetics of metronidazole and tinidazole in man. Arzneimittel-Forsch., 22, 2128-2132. Whelan, J. P. F. and Hale, J. H. (1973). Bactericidal activity of metronidazole against Bacteroidesfragilis. J. clin. Path., 26, 393-395.
A comparison of the in vitro activity of metronidazole, tinidazole, and nimorazole against Gram-negative anaerobic bacilli A. V. REYNOLDS, J. M. T. HAMILTON-MILLER, AND W. BRUMFITT From the Royal Free Hospital, Department of Medical Microbiology, Pond Street, Hampstead, London NW3 2QG
The in vitro activities of metronidazole, nimorazole, and tinidazole were compared against 69 strains of obligately anaerobic Gram-negative bacilli. Geometric mean MICs were 0 34, 1 05, and 0-28 Hg/ml respectively. Thirty-six strains were also tested by the disk method. Correlation between MIC and diameter of the zones of inhibition was poor.
SYNOPSIS
Metronidazole (1-,B-hydroxymethyl-2-methyl-5nitroimidazole, Flagyl, May & Baker) is at present the drug used most widely for the treatment of Trichomonas vaginalis infection. It is of interest that it has also been effective in the treatment of acute ulcerative gingivitis (Shinn, 1962; Davies et al, 1964; Shinn et al, 1965), an infection in which anaerobic bacteria (Treponema vincentii and Fusobacterium spp.) are believed to have an aetiological r6le. Recent studies have shown that all the clinically important species of anaerobic bacteria are sensitive to therapeutically attainable concentrations of metronidazole (Fuzi and Csukas, 1969a, b, c, 1970; Prince et al, 1969; Ueno et al, 1971; Videau, 1971; Nastro and Finegold, 1972; Tally et al, 1972, 1974; Finegold and Sutter, 1972; Whelan and Hale, 1973; Mitchell, 1973; Washington et al, 1974; Dornbusch and Nord, 1974; Phillips, 1974; Staneck and Washington, 1974). In addition, metronidazole is consistently bactericidal in action, which is a property not shown by clindamycin, rifampicin or chloramphenicol (Nastro and Finegold, 1972; Finegold and Sutter, 1972; Whelan and Hale, 1973). Two reports have appeared on the clinical use of metronidazole: Tally et al (1972) successfully treated three patients with infections caused by Gramnegative anaerobic bacteria, and the Study Group (1974) showed that metronidazole was effective both therapeutically and prophylactically in gynaecological patients with pelvic sepsis. Nimorazole1 (1-(N-P-ethylmorpholino)-5-nitro"Previously known as nitrinidazine. Received for publication 14 April 1975.
imidazole, Nagoxin, Carlo Erba) and tinidazole (1
(2-ethylsulphonylethyl)-2-methyl-5-nitroimidazole Fasigyn, Pfizer) are also 5-nitroimidazoles whose antitrichomonal activity in vitro is similar to that of metronidazole (de Carneri et al, 1969; Howes et al, 1970). Both have been used successfully to treat trichomoniasis, with reduction in both daily doses and treatment period compared with metronidazole (Rognoni and Sideri, 1969; Gyorik and Wenner, 1971). Nimorazole was also found to be as effective as metronidazole in the treatment of acute ulcerative gingivitis (Lozdan et al, 1971). Considering the superior pharmacokinetic properties of nimorazole and tinidazole, we thought it worthwhile to carry out a comparison of the in vitro activity of these three 5-nitroimidazoles against Gram-negative anaerobic bacteria. Methods and Materials
5-NITROIMIDAZOLES Pure samples of metronidazole, nimorazole, and tinidazole were supplied by May & Baker, Carlo Erba, and Pfizer, respectively. BACTERIAL STRAINS
Sixty-nine strains were studied: all were obligately anaerobic Gram-negative rods. Twenty-four strains (8 Fusobacterium spp. and 16 Bacteroides fragilis) were either reference strains (from the National Collection of Type Cultures, London or Center for Disease Control, Ga, USA) or identified clinical isolates from other hospitals. The majority of the remaining 45 strains had been isolated from clinical 775
A. V. Reynolds, J. M. T. Hamilton-Miller, and W. Brumfitt
776
material from the Royal Free Hospital but some were kindly donated by other hospitals. We provisionally identified 43 of these as B. fragilis, since they were resistant to polymyxin B (10 ,ug/ml), kanamycin (2 mg/ml), and benzylpenicillin (2 u/ml) (Finegold et al, 1967). The other two strains were sensitive to benzylpenicillin and thus were classified as Bacteroides spp. MIC DETERMINATION
Throughout these experiments, media were supplemented with yeast extract (1 %), haemin (5 ,ug/ml), and menadione (0 5 ,ug/ml). Serial doubling dilutions of the drugs were made in brain heart infusion agar (Oxoid, CM 375) containing 5 % lysed, defibrinated horse blood. Bacteria were grown in 10 ml amounts of thioglycollate broth (Baltimore Biological Laboratories, no. 135C) incubated anaerobically at 37°C for 22 hr, using the GasPak system. Such cultures were diluted 1:500 in sterile 0-06 M sodium phosphate buffer containing 0 3 g/l cysteine hydrochloride, and inoculated on to the agar with a multiple inoculating device (see Reynolds et al, 1974). Plates were incubated at 37°C for 42 hr using the GasPak system. Control plates containing no antibiotic were inoculated and incubated both aerobically and anaerobically. MIC was taken as the lowest drug concentration that completely inhibited growth or MIC
allowed the development of no more than two discrete colonies; such a finding is evidence of at least 99 95 % inhibition. Disk Diffusion Tests
Broth cultures were grown as above and diluted 1: 50 in buffer. Fifteen millilitre amounts of supplemented (see above) brain heart infusion agar + 5 % whole horse blood in 9 cm plates were flooded with the diluted cultures and excess fluid was removed. Disks containing 2f5 ,ug of drug were prepared by dipping 6 mm AA disks (Whatman) into solutions of each individual drug which contained 125 ,g/ml. These were then placed onto the surface of the agar. Only one disk was put on to each plate. Plates were incubated anaerobically using the GasPak system for 18 hr at 37°C, and zones of inhibition were measured with calipers. Analysis of Data Graphs were constructed to compare zone size with MIC for each compound; in addition, results for all three compounds were combined onto a single plot. Analysis was as described by Ericsson and Sherris (1971). Results
As
can
be
seen
from table I, there
was no
(Mg/ml) 2-0
0 062
0 125
0-25
05
1
2 3
4 27 1
24 1
3
7
2 41 2
5 6
1 15 1
3
1
1.0
40
8-0
Metronidazole
Fusobacterium sp B. fragilis Bacteroides sp Nimorazole Fusobacterium sp B. fragilis Bacteroides sp Tinidazole Fusobacterium sp B. fragilis Bacteroides sp
5 31 1 Metronidazole 0 34 ug/ml Nimorazole 1-05 Mg/ml 0-28 pg/ml Tinidazole 2 7
2
Geometric mean MIC
1 4
1 1 1
Table I Activity of metronidazole, nimorazole, and tinidazole against Gram-negative anaerobic bacteria Nitroimidazole
Fitted Line y a
Metromidazole
-2-23
Nimorazole Tinidazole
- 2-84 - 162 - 2-90
Composite
=
Correlation Coefficient
SE of Estimate
0 59 0-63 0 53 0 75
3-33 3-62 2 71 3 49
ax + b
b 41-45 43-67 36-46 45-41
Table II Correlation of MIC to zone diameter
1
clear
Nitroimidazoles against Gram-negative anaerobic bacilli
777
difference between the sensitivity patterns of Fusobacterium spp and B. fragilis; therefore it was considered to be justified to combine data for all strains. Tinidazole was slightly more active than was metronidazole, the two compounds having geometric mean MICs of 0-28 and 0 34 ,ug/ml respectively. Nimorazole was about threefold less active, having a geometric mean MIC of 1 05 .tg/ml. Thirty-six strains were also tested by the disk method, and the results are summarized in table IL. To increase the spread of values and thereby make the interpretation of results easier a composite scatter diagram was constructed and a statistical analysis of this is also shown in table II.
nitrofurans. The possession of such a group may be the determining factor in such activity, and it will be of interest to test this hypothesis with further compounds. The nitroimidazoles are at present available only as oral preparations, which may be a disadvantage if the treatment of critically ill patients is contemplated. In view of the fact that metronidazole has been used successfully on a small scale for the treatment of systemic bacterial infections, the superior pharmacokinetics (Welling and Monro, 1972; de Carneri, 1969) of the two other related compounds allied to their equivalent antimicrobial activity make clinical studies of tinidazole and nimorazole mandatory.
Discussion
We should like to thank Dr McFadzean, Dr Rooney, and Dr O'Neill East for the gifts of metronidazole, nimorazole, and tinidazole respectively. Doctors S. M. Finegold and V. L. Sutter, Wadsworth VA Hospital, Los Angeles, California, H. Beerens, University of Lille, France, M. Sebald, Pasteur Institute, Paris, France, G. Lombard, Center for Disease Control, Atlanta, Georgia, R. B. Sykes, Glaxo Research, Greenford, Middlesex, B. S. Drasar, St Mary's Hospital, London, and A. Percival, University of Liverpool, kindly provided strains of Gram-negative anaerobes.
A large number of reports over the past few years have shown metronidazole to be highly active against both Gram-positive and Gram-negative anaerobic bacteria (for review see Hamilton-Miller, 1975). Two other nitroimidazoles, nimorazole and tinidazole, are currently available for the treatment of trichomoniasis (although tinidazole is not at present available on prescription in the UK.) We have not seen any reports on the antibacterial activity of nimorazole, and only two concerning tinidazole. Edwards et al (1973) reported that tinidazole was slightly less active than metronidazole against eight strains of Clostridium spp, a conclusion confirmed by Dornbusch and Nord (1974), who also found tinidazole to be more active than metronidazole against Gram-negative organisms. Our results have been extended to include nimorazole but in terms of metronidazole and tinidazole they agree very closely with those of Dornbusch and Nord (1974). Summation of published data available on the activity of metronidazole against anaerobic bacteria shows that 1052 out of 1071 strains were sensitive (taking the break-point as being either 12-5 or 16 ,ug/ml). Poor correlation was found in the disk test between zone size and sensitivity. However, since all the strains tested here may be regarded as sensitive, criteria allowing precise interpretation cannot be stated. We suggest an MIC of 16 ,ug/ml as the breakpoint, since this value should be obtainable systemically after a 1 g dose of any of these three compounds. If this value is accepted, strains giving a zone size greater than 13 mm for metronidazole, 7 mm for nimorazole, and 15 mm for tinidazole can be considered as sensitive. Three groups of compounds containing a primary nitro group among other features have now been shown to possess good activity against anaerobic bacteria, chloramphenicol, nitroimidazoles, and
References
Davies, A. H., McFadzean, J. A., and Squires, S. (1964). Treatment of Vincent's stomatitis with metronidazole. Brit. med. J., 1, 1149-1150. de Carneri, 1. (1969). Antiprotozoan activity of nitroimidazoles. Arzneimittel Forsch., 19, 382-386. de Carneri, I., Cantone, A., Emaneuli, A., Giraldi, P. N., Logemann, W., Meinardi, G., Monti, G., Nannini, G., Tosolini, G., and Vita, G. (1969). Nitrimidazine: a new systemic trichomonacide. 6th International Congress of Chemotherapy, Tokyo (Kyoto). Dornbusch, K. and Nord, C. E. (1974). In vitro effect of metronidazole and tinidazole on anaerobic bacteria. Med. microbiol. Immunol., 160, 265-267. Edwards, D. I., Dye, M., and Came, H. (1973). The selective toxicity of antimicrobial nitroheterocyclic drugs. J. gen. Microbiol., 76, 135-145. Ericsson, H. M. and Sherris, J. C. (1971). Antibiotic sensitivity testing. Report of an international collaborative study. Acta. path. microbiol. scand. Section B., Suppl. no. 217.
Finegold, S. M., Harada, N. E., and Miller, L. G. (1967). Antibiotic susceptibility patterns as aids in classification and characterization of gram-negative anaerobic bacilli. J. Bact., 94, 1443-1450. Finegold, S. M. and Sutter, V. L. (1972). Antimicrobial susceptibility of anaerobic Gram-negative bacilli. In Host Resistance to Commensal Bacteria, edited by T. MacPhee. Churchill Livingstone, Edinburgh. Fuzi, M. and Csukas, Z. (1969a). A metronidazol egy eddig ismeretlen antibacterialis hatasa. Orv. Hetil., 110, 16051606.
Fuzi, M. and Csukas, Z. (1969b). A szAjfl6ra mikroorganiz-
778
A. V. Reynolds, J. M. T. Hamilton-Miller, and W. Brumfitt
musainak metronidazol erzekenysege. Orv. Hetil., 110, 2154:2155. Fuzi, M. and Csukas, Z. (1969c). Veillonella-tdrzsek metronidazol-erzekenysege. Fogorv. Szle., 62, 324-326. Fuzi, M. and Csukas, Z. (1970). Das antibacterielle Wirkungs-spektrum des Metronidazols. Zbl. Bakt. I. Abt. Orig., 213, 258-262. Gyorik, W. and Wenner, R. (1971). Therapie der Trichomonaden-infektion mit Tinidazol im Vergleich zu Metronidazol. Schweiz. Rundschau. Med. (PRAXIS), 60, 1612-1614. Hamilton-Miller, J. M. T. (1975). Antimicrobial agents active against anaerobes. J. antimicrob. Chemother., 1 (in press). Howes, H. L. Jr., Lynch, J. E., and Kivlin, J. L. (1970). Tinidazole, a new antiprotozoal agent: effect on Trichomonas and other protozoa. Antimicrobial Agents and Chemotherapy-1969, 8, 261-266. Lozdan, J., Sheiham, A., Pearlman, B. A., Keiser, B., Rachanis, C. C., and Meyer, R. (1971). The use of nitrimidazine in the treatment of acute ulcerative gingivitis. A double-blind controlled trial. Brit. dent. J., 130, 294-296. Mitchell, A. A. B. (1973). Incidence and isolation of Bacteroides species from clinical material and their sensitivity to antibiotics. J. clin. Path., 26, 738-741. Nastro, L. J. and Finegold, S. M. (1972). Bactericidal activity of five antimicrobial agents against Bacteroides fragilis. J. infect. Dis., 126, 104-107. Phillips, 1. (1974). Antibiotic sensitivity of non-sporing anaerobes. In Infections with Non-Sporing Anaerobic Bacteria, edited by I. Phillips, and M. Sussman. Churchill, Livingstone, Edinburgh. Prince, H. M., Grunberg, E., Titsworth, E., and deLorenzo, W. F. (1969). Effects of 1-(2-nitro-1-imidazolyl)-3-methoxy -2-propanol and 2-methyl-5-nitroimidazole-1-ethanol against anaerobic and aerobic bacteria and protozoa. Appl. Microbiol., 18, 728-730. Reynolds, A. V., Hamilton-Miller, J. M. T., and Brumfitt, W. (1974). Newer aminoglycosides-amikacin and tobramycin: an in-vitro comparison with kanamycin and gentamicin. Brit. med. J., 3, 778-780.
Rognoni, V. and Sideri, L. (1969). Treatment of vaginitis with particular reference to those forms due to a protozoan causative factor. Clinical results with a new nitroimidazole preparation. Riv. Ostet. Ginec., 51, 237-246. Shinn, D. L. S. (1962). Metronidazole in acute ulcerative gingivitis. Lancet, 1, 1191. Shinn, D. L. S., Squires, S., and McFadzean, J. A. (1965). The treatment of Vincent's disease with metronidazole. Dent. Practit. dent. Rec., 15, 275-280. Staneck, J. L. and Washington, J. A. 1I. (1974). Antimicrobial susceptibilities of anaerobic bacteria: recent clinical isolates. Antimicrob. Agents Chemother., 6, 311315. Study Group (1974). Metronidazole in the prevention and treatment of bacteroides infections in gynaecological patients. Lancet, 2, 1540-1543. Tally, F. P., Armfield, A. Y., Dowell, V. R., Jr., Kwok, Y-Y., Sutter, V. L., and Finegold, S. M. (1974). Susceptibility of Clostridiumramosum to antimicrobial agents. Antimicrob. Agents Chemother., 5, 589-593. Tally, F. P., Sutter, V. L., and Finegold, S. M. (1972). Metronidazole versus anaerobes. In vitro data and initial clinical observations. Calif. Med., 117, no. 6, 22-26. Ueno, K., Ninomiya, K., and Suzuki, S. (1971). Antibacterial activity of metronidazole against anaerobic bacteria. Chemotherapy (Japan), 19, 111-114. Videau, D. (1971). Association metronidazole-spiramycin sur les germes ana6robes. Path. Biol., 19, 661-666. Washington, J. A., II, Martin, W. J., and Hermans, P. E. (1974). In vitro susceptibility of anaerobic bacteria isolated from blood cultures. In Anaerobic Bacteria: Role in Disease, p. 427, edited by A. Balows, R. M. Dehaan, V. R. Dowell, Jr., and L. B. Guze. Thomas, Springfield, Illinois. Welling P. G. and Monro, A. M. (1972). The pharmacokinetics of metronidazole and tinidazole in man. Arzneimittel-Forsch., 22, 2128-2132. Whelan, J. P. F. and Hale, J. H. (1973). Bactericidal activity of metronidazole against Bacteroidesfragilis. J. clin. Path., 26, 393-395.
